【作者】 叶燎原； 刘本玉； 缪升；

【Author】 Ye Liaoyuan Liu Benyu Miao Sheng (Institute of Public Safety and Disaster Prevention,Yunnan University,Kunming 650091,P.R.China)

【机构】 云南大学公共安全与区域防灾研究中心；

【摘要】 本文综述了目前国内外应用人工神经网络方法来进行防震减灾问题的分析处理。人工神经网络中的 BP 网络在防震减灾的应用最为广泛,其余的有 RBF 网络、自组织神经网络、模糊神经网络、遗传算法等也有些应用。主要的应用领域有在确定地震参数中的应用、在潜在震源划分中的应用、在地震预报中的应用、在确定地震动参数中的应用、在场地液化评价中的应用以及在建筑物和次生火灾震害预测的中的应用等。最后,本文还对形成标准的人工神经网络方法的相关问题进行了讨论。更多还原

【Abstract】 The complex nature of the seismic-relate problem made the nonparametric methods like artificial intelli- gence quickly developed in this field.Neural networks are increasingly popular due to big progress in neural net- work modeling techniques and quick advancement of personal computer.Back propagation model is the most wide ly used,along with the self-organized networks model ,fuzzy neural networks,and probabilistic neural networks.In this paper,the applications of artificial neural networks in various branches of earthquake resistance and disaster prevention have been examined here.These fields have a lot of history data,such as determination of seismic pa- rameters,potential seismic source zoning,earthquake prediction,the evaluation of site liquefaction,buildings and in- frastructure damage prediction,and seismic induced disaster prediction etc.The key steps and points of forming normal Artificial Neural Networks method in these fields are discussed,and the integrated artificial neural network modeling is discussed for the nonlinear processes in these fields.更多还原